br Methods br Results br Discussion

Methods
Results
Discussion Our results show that the DRG and the spinal cord express significant amount of GPR35, a previously orphan Gi protein coupled receptor whose stimulation with zaprinast or with KYNA may reduce inflammatory pain. The receptor is also expressed in primary cultures of glial cells and has a rather high affinity for zaprinast and KYNA (EC50 0.3 μM). It has been reported that GPR35 stimulation may inhibit N-type calcium channel controlling cellular excitability and transmitter release and our results are certainly in line with these observations (Guo et al., 2008). We previously demonstrated that nanomolar concentrations of KYNA may effectively reduce glutamate release and glutamate extracellular concentrations in various DAPI hydrochloride regions (Carpenedo et al., 2001) and similar results have been obtained with zaprinast (Cozzi et al., preliminary unpublished observations). GPR35 seems therefore able to regulate excitatory transmitter release, an action that could explain the analgesic effects here reported. Nanomolar concentrations of KYNA also inhibits the heat-shock-induced FGFα release from non-neuronal cell lines (Di Serio et al., 2005) and neurotrophin release in damaged or inflamed tissues is considered an important event in nociceptor activation (Pezet and McMahon, 2006). It is possible to propose that inhibition of neurotrophin release may contribute to the potent analgesic action of zaprinast and KYNA. In order to increase KYNA levels in the brain we used two strategies: the first consisted in the administration of l-kynurenine, with the aim of increasing KYNA synthesis, the second consisted in the administration of probenecid to reduce KYNA disposal. Both strategies resulted in a significant reduction of writes. l-kynurenine readily crosses the blood brain barrier and may be converted into KYNA by kynurenine aminotransferases (KATs). KATs are enzymes that are abundantly expressed in astrocytes in both the brain and spinal cord (Du et al., 1992, Guidetti et al., 1997, Kapoor et al., 1997). Since their affinity for the substrate is rather low (in the millimolar range), KYNA synthesis is a direct function of kynurenine availability (Okuno et al., 1991, Guidetti et al., 1997, Kapoor et al., 1997). KYNA effects in the central nervous system have been ascribed to the antagonism of the glycine site of NMDA, and NMDA receptor antagonists are known to attenuate the pain in various animal models (Chen et al., 2009). However, endogenous concentrations of KYNA in the brain and spinal cord are in the nanomolar range while the concentrations that are required to antagonize the glycine site of NMDA receptors are higher than 30 μmolar (Pellegrini-Giampietro et al., 1989, Moroni et al., 1989). A number of observations indicate that nanomolar concentrations of KYNA are sufficient to reduce brain excitatory transmission and to protect against excitotoxic neuronal damage (Bacciottini et al., 1987, Russi et al., 1989, Moroni et al., 1991, Vécsei et al., 1991, Russi et al., 1992, Nozaki and Beal, 1992, Carpenedo et al., 1994, Miranda et al., 1997, Erhardt et al., 2000), possibly KYNA effects in the brain are at least partially mediated by GPR35activation and reduced glutamate release. Indeed, our data shows (see Fig. 2) that the affinity of KYNA at native mouse GPR35 receptors is actually quite high and that KYNA EC50s is in the submicromolar range. It has been reported that kynurenine may have analgesic actions in the tail flick pain model (Heyliger et al., 1998) and our data confirm these observation and show that analgesia is particularly evident in the inflammatory pain (Collier et al., 1968). Since probenecid has effects similar to those of kynurenine, our data strongly suggest that KYNA is the mediator of kynurenine actions. GPR35 involvement in this analgesic action is consistent with the effects of low doses of zaprinast, a high affinity agonist for GPR35 (Taniguchi et al., 2006) and by the expression of GPR35 in both the dorsal root ganglia and the spinal cord (see Fig. 1). The possibility that zaprinast and KYNA analgesia is mediated by the same mechanism is in agreement with the observation that maximal doses of the two agents had no additive action (see Fig. 4B). Our results are also in line with a recent report by (Zhao et al., 2010) showing that pamoic acid, a newly identified GPR35 agonist, is also provided with antinociceptive actions in the acetic acid-induced pain model here described. CID2745687, a new and selective GPR35 antagonist, prevented pamoic acid effects “in vitro”. Preliminary observations from our laboratory confirmed pamoic acid analgesic action.